$r$-Process Nucleosynthesis from Three-dimensional Jet-driven Core-Collapse Supernovae with Magnetic Misalignments

We investigate $r$-process nucleosynthesis in three-dimensional general relativistic magnetohydrodynamic simulations of jet-driven supernovae resulting from rapidly rotating, strongly magnetized core-collapse. We explore the effect of misaligning the pre-collapse magnetic field with respect to the rotation axis by performing four simulations: one aligned model and models with 15, 30, and 45 degree misalignments. The simulations we present employ a microphysical finite-temperature equation of state and a leakage scheme that captures the overall energetics and lepton number exchange due to post-bounce neutrino emission and absorption. We track the thermodynamic properties of the ejected material with Lagrangian tracer particles and analyse its composition with the nuclear reaction network SkyNet. By using different neutrino luminosities in post-processing the tracer data with SkyNet, we constrain the impact of uncertainties in neutrino luminosities. We find that, for the aligned model considered here, the use of an approximate leakage scheme results in neutrino luminosity uncertainties corresponding to a factor of 100-1000 uncertainty in the abundance of third peak $r$-process elements. Our results show that for misalignments of 30 degrees or less, $r$-process elements are robustly produced as long as neutrino luminosities are reasonably low ($\lesssim 5 \times 10^{52}$ erg s$^{-1}$). For a more extreme misalignment of 45 degrees, we find the production of $r$-process elements beyond the second peak significantly reduced. We conclude that robust $r$-process nucleosynthesis in magnetorotational supernovae requires a progenitor stellar core with a large poloidal magnetic field component that is at least moderately (within $\sim 30$ degrees) aligned with the rotation axis.